Experience in Scaling-Up of Photo-Thermo-Catalytic Purification of Process Gasses from NO_x

Abstract

Photo-thermo-catalytic or PTC purification of process gasses (i.e., air, flue gases, and others) from NO_x is presented in this study. A discussion of temperature’s role in photocatalytic NO_x removal and the progress of photo-thermo-catalytic reactors for the NO_x removal process are presented. Lab- and pilot-scale reactors are described. The impact of temperature on the photocatalytic conversion of hydrocarbons is analyzed with regard to its relation to the photocatalytic selective reduction of NO_x (photo-SCR). Another important issue is light transfer in pilot-scale reactors due to the sensitivity of light sources to temperature. Examples of light transfer solutions in photo-thermo-catalytic reactors are presented. Finally, the further development of photo-thermo-catalytic reactors is discussed, including pressurized systems.

1. Introduction

After SO₂, CO, PM, and VOC, nitrogen oxides (NO_x, i.e., NO and NO₂) belong to the main atmospheric pollutants in existence [1]. It is known that the main sources of NO_x emissions are stationary combustion chambers and mobile chambers (e.g., vehicle engines, aircraft engines, and ship engines) [2]. There are many methods of NO_x removal; nevertheless, they can be classified into two main groups, namely, primary and secondary methods. Within the second group, photocatalytic methods of NO_x removal are of special interest due to the potential application of sunlight in the process of air purification [3]. Photo-thermal catalysis or PTC has recently emerged as a new way to enhance chemical reactions. The main benefits of PTC can be obtained from the combination of thermal and photocatalytic processes. It has been emphasized that more efficient utilization of sunlight as an energy source can be obtained. Thus, more effective harvesting of the solar spectrum can be observed, as low-energy visible and infrared photons are insufficient to promote photocatalytic reactions [4]. Ma and co-workers [5] presented a definition of the synergy effect due to the enhancement of photocatalytic (PC) and thermal catalytic (TC) processes. According to these researchers, the synergistic effects are possible due to cooperative utilization between light and heat in one reaction system. Thus, the enhancement of the catalytic reaction can be observed, and the catalytic effect is higher than the arithmetic sum of PC and TC. The authors described four possible effects of synergy, namely: (1) enhancing catalytic activity, (2) enhancing selectivity for the target product, (3) decreasing the reaction temperature, such as the photothermal effect of the Group VIII nanocatalysts, to eliminate external heating, and (4) enhancing catalyst durability and lifetime. Based on these categories, Ma and co-workers [5] proposed four types of photo-thermal synergistic modes, namely, thermal-assisted photocatalysis (TAPC), photo-assisted thermocatalysis, (PATC), photo-driven thermocatalysis (PDTC), and photothermal co-catalysis (PTCC). TAPC is defined as light being the main driving force behind the reaction. Heat assists the photocatalytic process rather than driving TC. PATC is defined as heat being the main driving force behind the reaction. Light promotes the thermocatalytic process rather than driving PC. The light could contribute to the enhancement of TC activities in several ways, including increasing the local temperature at the catalyst’s surface by electron relaxation and non-radiative recombination. PDTC is defined as photo-induced heat being the main driving force for the reaction, with light indirectly driving TC rather than directly driving PC. PTCC is defined as both light and heat being the driving forces for the reaction, with light directly driving PC and heat driving TC, giving a better synergistic performance than a combination of PC and TC [5].

Descriptions of photo-thermal catalysis, including its mechanisms, synergy effects, types of reactors, and materials, have been presented in review papers by Ma and colleagues [5], Keller and colleagues [6], Tang and colleagues [7], and Kho and colleagues [8]. In these works, much fundamental and useful information can be found to classify PTC modes and configurations. For example, some information about VOC degradation presented by Ma and colleagues [5] can be adopted in the process of NO_x removal due to the references to photo-selective catalytic reduction (photo-SCR). Keller and colleagues [6] presented different configurations of PTC reactors. Nevertheless, these research groups have not presented an analysis of PTC in terms of NO_x removal. They have focused on such processes as solar fuel generation by H₂ generation [5,6], CO₂ conversion (by methanation (Sabatier reaction) CO₂ + 4H₂ → CH₄ + H₂O, reverse water gas shift reaction CO₂ + H₂ → CO + H₂O, dry reforming reaction CO₂ + CH₄ → 2H₂ + 2CO for Fisher–Tropsch synthesis) [5,6,8], the degradation of organic pollutants (VOC and dyes) [5], and organic synthesis [5]. PTC can provide some beneficial effects for photocatalytic NO_x removal (especially photo-selective catalytic reduction). To the best of our knowledge, photo-thermo catalysis in terms of deNOx processes has not been reviewed before. This paper aims to present the progress of photo-thermo-catalytic NO_x removal using lab- and pilot-scale facilities. Photocatalytic NO_x removal is currently an intensively investigated area that is included under crucial environmental issues [3,9,10,11,12,13,14].

2. PTC NO_x Removal: State of the Art

2.1. The Role of Temperature in the Solution to the “Water Problem”

There are three types of photocatalytic removal of NO_x: (i) photo-selective catalytic reduction (photo-SCR), (ii) photo-oxidation, and (iii) photo-decomposition. Photo-decomposition and photo-SCR are reduction processes. The main objective of these deNOx processes is to convert these pollutants to N₂ and other harmless compounds. The photo-oxidation of NO_x results in the formation of nitric acid, which must be removed from the photocatalyst’s surface [3]. The problem with photo-NO_x-decomposition is the low productivity of the process, which often yields N₂O and not N₂ as the primary product [11].

The necessity of a combination of thermal and photocatalytic modes in the NO_x removal processes was acknowledged when selective photocatalytic reduction (photo-SCR) met a serious barrier. Namely, the existence of oxidative compounds such as H₂O and O₂ in the reaction system significantly reduced the efficiency of photo-SCR. Indeed, this problem significantly inhibited the development of photo-SCR, because oxidative compounds (O₂, H₂O, CO₂) exist in flue gas. Thus, the reduction of NO_x from real-scale flue gas seemed to be rather impossible.

This problem came from the fact that the surface of TiO₂ becomes superhydrophilic under UV light irradiation. A very thin layer of water is created on the TiO₂ surface under UV-light irradiation, which inhibits photo-SCR. This water layer inhibits the transport of reactants and products from the photocatalyst surface into the gas zone, and vice versa. Additionally, water as an oxidizer enhances the formation of NO₂, which is the typical product of photo-oxidation [3]. This negative effect has been confirmed for O₂ by Yu and colleagues [15], who noticed that the conversion of NO was decreased to less than 10% in the presence of water vapor and oxygen. Another reason for the reduction in NO conversion is the competitive adsorption of H₂O and reducing species. Namely, Nguyen and colleagues reported [11] that the selectivity and efficiency of the conversion from NO to N₂ were reduced when H₂O was introduced to the reaction system because of the competitive adsorption of H₂O and NH₃ on active sites. Nevertheless, temperature also influences the adsorption of NO_x [16].

The first mention of a possible solution to this impasse or deadlock came from the research results presented by Poulston and colleagues in 2009 [17]. They observed that higher temperatures (150 °C) promoted the transformation of NO into N₂ and N₂O under UV irradiation despite the presence of oxygen (12%) in a gaseous atmosphere. They tested TiO₂ as a photocatalyst and several hydrocarbons (C₄H₁₀, C₃H₈, C₃H₆, C₂H₆, and C₂H₄) as reducing agents. It was observed that NO was mostly transformed into NO₂ when the process temperature was lower than 42 °C [3,17]. It should be mentioned that Poulston and colleagues [17] investigated photo-oxidation of ethane, ethene, propane, and propene over titania, and they considered NO to be an oxidation agent. Thus, the relation to photo-SCR did not directly reveal itself. Considering this phenomenon, in 2010, Lasek and Wu [18,19] noticed a chance to develop photo-SCR and solve the “water problem”. The first confirmation of the positive effect of a temperature increase on photo-SCR was presented using a monolith photoreactor [18] as well as a small-scale photo-thermal rector [19]. Next, the positive effect of temperature on photo-SCR was confirmed using different catalysts and reactors [20,21,22].

The first systematic investigations into the beneficial effect of temperature on photo-SCR were carried out using a lab-scale facility. Namely, photocatalytic reduction of nitric oxide was studied on a PtO_xPdO_y/TiO₂-coated monolith photoreactor using propane as the reducing agent. The photocatalytic reduction of NO in NO/propane and NO/propane/O₂ systems was studied by in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) [18].

The most important conclusion in terms of temperature’s impact on photo-SCR is that a temperature increase inhibits the oxidizing effect of O₂ and H₂O due to the change in adsorption equilibrium on the catalyst’s surface. Yu and colleagues [18] explained that as the temperature increased to 120 °C, NO conversion increased significantly, whereas NO₂ formation decreased. In other words, more NO was reduced, and less was oxidized. Such an effect was explained by the competition between O₂ and NO to capture the photogenerated electrons on the catalyst surface. It is known that NO is a polarized molecule and O₂ is not. Thus, the partially negative O atom on NO has a stronger affinity for the Ti⁴⁺ site to which the photo-generated electrons migrate. Note that the basic heterogeneous photocatalytic process involves the creation of electron and hole pairs during excitation under UV or visible light irradiation. At higher temperatures (120 °C), the concentration of adsorbed NO and O₂ on the catalyst surface decreases. However, NO exhibits stronger adsorption capability on the catalyst surface than O₂, resulting in a higher NO conversion of close to 73%, as in the NO/C₃H₈ system. The adsorption of oxygen is relatively sensitive to temperature. As the temperature increases, less O₂ is adsorbed and less nitrate, nitrite, and hydrocarbon intermediates are formed. Similar explanations have been given for the presence of H₂O in the reaction system. As the temperature increases to 120 °C, a large number of H₂O molecules desorb, releasing the active sites from the catalyst surface. The adsorption capacity of propane and NO is not sensitive in this temperature range. As a result, the ratio of adsorbed NO to propane increases, thereby increasing the NO removal at 120 °C. Lasek and co-workers [19] investigated photo-SCR of NO in the presence of water. The experiments were carried out in a continuous-flow photoreactor with 0.55–1.6 v% water at 30–120 °C under a UV-light intensity of ∼200 mW/cm². The C₃H₈/NO molar ratio in the feed ranged from 0.8–16.8 at a volume hourly space velocity (VHSV) from 330–1090 h⁻¹. An increase in temperature at a suitable C₃H₈/NO ratio can minimize NO₂ formation (see Figure 1). Increasing temperature inhibited NO₂ formation in the presence of water when the PdO/TiO₂ photocatalyst was applied in the photo-SCR process.

The beneficial effect of temperature on photo-SRC was confirmed several times by the research group supervised by J.C.S. Wu (Taiwanese side) and J. Lasek (Polish side) [3,19,20,22]. A revision of the obtained results is presented in

Table 1. Process of thermal-assisted photo-SCR of NOx using lab- and pilot-scale photo-reactors.

No	Catalyst	Reactor	Conditions	Max. NOx Conversion, %	Ref.
1	Powder form, Pd/PdO-loaded TiO2 (Degussa P25; 75% anatase and 25% rutile) specific surface area of 50 m2/g	Lab-scale photo-reactor aluminum alloy 6061, quartz window. The light-exposed area is 273 mm2, and the shape of the reactor is a double isosceles triangles with cut vertex (length of 3 to 5 mm, width of 12 mm, and depth of 1 mm), and the internal volume of the reactor is 0.546 mL. The photoreactor inserted with a thermocouple was settled on a heating plate to precisely control the reaction temperature.	400 ppm NO, 5000 ppm C3H8, and pure N2 as water vapor carrier (0–1.6 v.%). The light intensity and wavelength are ~200mW/cm2 and 365 nm, respectively. Flow rate 3–10 mL/min (equivalent to volume hourly space velocity (GHSV) from 330–1090 h−1), temperature 30–120 °C.	Dry 94% Humid 90%	[19]
2	PtOxPdOy/TiO2-coated monolith (4.2 cm diameter and 6.0 cm length)	Lab-scale monolith photoreactor. The catalyst-coated monolith was inserted by side-glowed optical fiber in each channel and placed inside a circular quartz vessel.	400 ppm NO, 5000 ppm C3H8, O2 5 v.%, H2O (water saturator at room temperature), N2 UV-A light (320–500 nm) at intensity of 46.9 mW/cm2, flow rate 7 mL/min (GHSV 3.15 h–1), temperature 25, 70, and 120 °C.	Dry 90% Humid 65% O2 67.1%	[18]
3	TiO2 supported on a spherical α-Al2O3/γ-Al2O3	Modified lab-scale photoreactor (no. 1 in this table): a cylindrical notch of 15 mm in diameter and 1 mm in height, with exposure area of 176.63 mm2. The photoreactor inserted with a thermocouple was settled on a heating plate to precisely control the reaction temperature.	Lab-scale test 400 ppmv NO; 2000 ppmv C4H10; 4 v% O2; 4 v% H2O; and balancing by N2. The total feed gas volumetric flow rate was 25 mL/min (GHSV 10,000 h−1). The light source was provided by a mercury arc lamp (EXFO S1500, 200 mW cm−2, 320–500 nm), temperature 40–300 °C.	43%	[20]
4	TiO2 supported on a spherical α-Al2O3/γ-Al2O3	Pilot-scale photoreactor (annular type, diameter of 100 mm and 80 mm, length of 1120 mm, flow rate of 50–300 L/h of air)	Real-condition flue gas from the coal burning in a domestic furnace (25 kW), total volumetric flow rate 2300 mL/min (GHSV of 80 h−1), temperature 120–130 °C.	68–75%	[20]
5	TiO2 photocatalyst was prepared by sol-gel method (powder form)	Lab-scale photoreactor (like in No. 3)	400 ppmv NO, 2000 ppmv C4H10, 4 vol.% O2, 4 vol.% H2O, and balance gas (N2) with (GHSV of 10,000 h−1). Mercury arc lamp (OmniCure S1500, 200 mW cm−2, 320–500 nm), temperature 40–300 °C.	Humid, O2, 80%	[21]
6	Titania nanosheet photocatalysts with dominantly exposed (001) reactive facets	Lab-scale photoreactor (like in No. 3)	400 ppmv NO; 2000 ppmv C4H10; 4 v% O2; 4 v% H2O; and balancing by N2. The total feed gas volumetric flow rate was 25 mL min−1 (GHSV 10,000 h−1). Mercury arc lamp (OmniCure S1500, 200 mW cm−2, 320–500 nm), temperature 40–160 °C.	Humid, O2, 87%	[22]

. It can be noticed that the conversion of NO_x varies in the range of 43–90%, depending on the catalyst type and process parameters. It should be noted that one of the most important parameters is gas hourly space velocity or GHSV, calculated as the volumetric flow rate divided by the volume of the reactor. This parameter gives the approximate value of the residence time (or contact time) of gaseous compounds in the catalytic reactor. Thus, a GHSV value in the range of 330–1090 h⁻¹ is the equivalent of a residence time of approx. 3.3–11 s. A GHSV of value 10,000 h^–1 corresponds to a residence time of 0.36 s. Then, it is clear that the obtained NO_x conversion at higher values of GHSV gives potentially higher possibilities of flue gas purification. In other words, a shorter contact time provides a chance to increase the flue gas flow rate by photoreactor.

The beneficial effect of increasing temperature in terms of water desorption was observed by Fu and colleagues [23] during the photocatalytic oxidation of ethylene over TiO₂ and a Pt-TiO₂ photocatalyst under UV light irradiation. Similar results were reported by Falconer and Magrini-Bair [24], who investigated the photocatalytic oxidation of acetaldehyde on a Pt-TiO₂ photocatalyst under UV light irradiation. The contribution of photocatalytic oxidation was the maximum at 140 °C, where conversion was 2.8 times that at 24 °C.

Yu and colleagues [22] performed a comparison of photo-SCR for different catalysts, i.e., commercial Degussa P-25, a TiO₂ photocatalyst prepared by the sol-gel method, and titania nanosheet photocatalysts with dominantly exposed (001) reactive facets. The highest NO_x removal (i.e., 64%) at GHSV = 10,000 h^–1 and 120 °C was obtained for the titania nanosheet photocatalysts (see Figure 2). The NO_x removal was calculated from (1); thus, it favors photo-SCR, because the transformation into NO₂ (like in photo-oxidation processes [3]) does not provide an increase in NO_x removal. Please note that NO_x = NO + NO₂; thus, in such defined NO_x removal, only photo-reduction provides a beneficial effect.

$${\mathrm{N}\mathrm{O}}_{\mathrm{x}}\mathrm{r}\mathrm{e}\mathrm{m}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{l}=\left(\frac{{\mathrm{N}\mathrm{O}}_{\mathrm{x}\mathrm{i}\mathrm{n}\mathrm{l}\mathrm{e}\mathrm{t}}-{\mathrm{N}\mathrm{O}}_{\mathrm{x}\mathrm{o}\mathrm{u}\mathrm{t}\mathrm{l}\mathrm{e}\mathrm{t}}}{{\mathrm{N}\mathrm{O}}_{\mathrm{x}\mathrm{i}\mathrm{n}\mathrm{l}\mathrm{e}\mathrm{t}}}\right)\times 100\%$$

(1)

Yamamoto and colleagues [25] studied the effect of the reaction temperature on photo-SCR in the presence of NH₃ over a TiO₂ photocatalyst and under very high GHSV conditions (100,000 h⁻¹). They found that the reaction temperature had a significant effect on the photo-SCR activity. The maximum NO conversion was achieved at 160 °C (i.e., 84%), and it decreased with a further increase in temperature (i.e., 77% at 320 °C). They explained that the beneficial effect of the temperature increase during photo-SCR is related to the decomposition of NH₂NO intermediates as a rate-determining step at room temperature in the presence of excess O₂ gas (see Equation (2), where S is the active site). The conversion of NO increased as the reaction temperature increased from 80 °C to 160 °C, because the decomposition process of the NH₂NO intermediates was promoted by the temperature increase. This process occurred without photo irradiation, indicating that the reaction rate can be enhanced by increasing the reaction temperature. Further increases in temperature resulted in a decrease in NO conversion. Two reasons were proposed for this observation. One was the thermal deactivation of the catalysts, and the other was that an exothermic and equilibrium reaction in the elementary steps affected the overall reaction rate.

The authors experimentally demonstrated that the decrease in NO activity was not due to the deactivation of the catalysts by temperature impact. Thus, they confirmed the second hypothesis by theoretical and experimental considerations. They explained that the adsorption of NH₃ (see Equation (3)) is an exothermic and equilibrium reaction and confirmed experimentally that the amount of NH₃ adsorption decreased significantly with increasing temperature. The increase in temperature not only decreased the equilibrium constant of NH₃ adsorption but also decreased the maximum adsorption capacity. In conclusion, the overall reaction rate decreased because both the total number of active sites and the value of the equilibrium constant of NH₃ adsorption decreased as the temperature increased from 160 °C to 320 °C. Thus, Yamamoto and colleagues [25] concluded that three kinetic and thermodynamic parameters influenced the reaction rate of photo-SCR: (1) the rate constant of decomposition of NH₂NO intermediates, (2) the total number of active sites, and (3) the equilibrium constant of NH₃ adsorption. In the low-temperature range (80–160 °C), the rate constant of decomposition of NH₂NO intermediates was dominant. In the medium-temperature range (160–220 °C), the total number of active sites and the equilibrium constant of NH₃ adsorption, which decreases with increasing temperature, start to contribute to the reaction rate. In the middle-temperature range (220–320 °C), the contribution of active sites and the equilibrium constant to the overall reaction rate becomes more dominant than the rate of NH₂NO intermediate decomposition. The selectivity towards N₂ increased slightly with increasing reaction temperature, reaching 100% above 180 °C [25].

$$\mathrm{N}{\mathrm{H}}_2\mathrm{N}\mathrm{O}-{\mathrm{S}}^{\mathrm{*}}\rightleftarrows {\mathrm{N}}_2+{\mathrm{H}}_2\mathrm{O}+{\mathrm{S}}^{\mathrm{*}}$$

(2)

$$\mathrm{N}{\mathrm{H}}_3+\mathrm{S}\rightleftarrows \mathrm{N}{\mathrm{H}}_3-\mathrm{S}$$

(3)

Nevertheless, the increase in temperature can bring some undesired effects. Wu and colleagues [26] explained that reaction temperature has a dual effect on reaction efficiency by (1) enhancing the desorption of intermediates to obtain more active sites for the reaction and (2) reducing the adsorption of CH₄ by the photocatalyst. The last effect produces a decrease in the reaction efficiency. They found that during photo-SCR in the presence of CH₄ over a Pt/TiO₂ photocatalyst, the conversion of NO and the selectivity for NO₂ were highest at 25 °C, suggesting that a small amount of NO was oxidized, and the result of the photo-oxidation (i.e., HNO₃) occupied a catalyst surface, blocking a further reducing reaction. At 50 °C, the NO conversion is relatively low compared to the reaction at 25 °C and 100 °C. They explained that 50 °C is not high enough for the desorption of intermediates; therefore, the adsorption of CH₄ decreases rather than the desorption of intermediates increasing, resulting in the inhibition of photocatalytic activity. At 100 °C, the selectivity for NO₂ was lowest, showing preferential conditions for photo-SCR. Yamamoto and colleagues [25] noticed that an increase in temperature above 160 °C caused a decrease in photocatalytic NO conversion. The optimal range of temperature during the photocatalytic oxidation of hydrocarbons and a further decrease in the conversion rate was observed by other researchers [24,25,27,28,29,30,31]. Kang and co-workers [31] presented the results of photo-thermo catalytic oxidation of propane over a TiO₂-WO₃-supported platinum catalyst. These results are of special interest because C₃H₈ was recognized as an effective reducing agent for photo-SCR. The photo-thermo catalysis of a semiconductor-supported Pt catalyst (Pt/TiO₂-WO₃) was found to enhance the catalytic oxidation of C₃H₈ at low temperatures and at a high O₂/C₃H₈ ratio (volume ratio: 20). The peroxycarbonate (-OCO₃) was found to be the intermediate for this reaction by in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS). Moreover, Kang and co-workers [31] noticed that when raising the temperature, the decomposition of the peroxycarbonate can be accelerated. The results obtained by Falconer and Magrini-Bair [24] indicated a similar tendency during the photocatalytic oxidation of a Pt-TiO₂ photocatalyst under UV light irradiation. The maximum photocatalytic rate of acetaldehyde oxidation was obtained at 140 °C, and a further increase in temperature caused a weak photocatalytic contribution. Wu and colleagues [28] reported that the optimum temperature during the photocatalytic oxidation of benzene over Degussa P-25 TiO₂ and under UV light irradiation was in the range of 160–180 °C. They explained that the change in benzene was caused by a change in the rate-limiting step (i.e., the chemical reaction rate and the adsorption/desorption rate). An increase in temperature causes the enhancement of the chemical reaction rate and product desorption. At the same time, the reactant adsorption rate is reduced. As a result, the chemical reaction rate or product desorption rate gradually exceeds the reactant adsorption rate, which gradually becomes the slowest step. Moreover, the adsorption of the reactant becomes the rate-limiting step against the chemical reaction or product desorption [28]. The influence of temperature on adsorption/desorption and chemical reaction rate was also underscored in the process of photo-catalytic NO_x conversion [11,15]. Kim and colleagues [32] reported the beneficial effect of temperature on photoinduced charge transfer during photocatalytic H₂ production. The impact of temperature on the photocatalytic oxidation of hydrocarbons is important in terms of NO_x conversion, because hydrocarbons can be considered as potential reducing agents during photo-SCR. A summary of the impact of temperature on the photocatalytic oxidation of selected hydrocarbons is presented in

Table 2. The optimal temperature range in the photocatalytic oxidation of selected hydrocarbons.

Reactant	Optimal Temperature Range, °C	Photocatalyst, Light	Ref.
Acetaldehyde	140	Pt-TiO2, UV (6 W black light, 0.08–0.10 mW/cm2)	[24]
Acetone	100	TiO2, 1000 W xenon lamp (313 nm interference filter, 7 mW/cm2)	[27]
Benzene	160–180	TiO2 (Degussa P-25), 15 W near-UV lamp (350 nm)	[28]

.

A crucial milestone of photo-thermo-catalytic NO_x removal was the application of this technique to real-condition flue gas. Yu and colleagues [20] presented the results of photo-SCR from real-condition flue gas taken from the combustion of coal in a domestic-scale boiler (25 kW_th). The results are presented in Figure 3 The temperature inside the reactor was maintained at 120 °C. The NO removal at the pilot scale was about 68–75% during the first run with a fresh catalyst. During the second run, the efficiency loss was observed due to the undesired effect of sulfates on the catalyst surface. This effect was observed due to the presence of sulfur in the coal matrix. It is known that the main product of sulfur’s oxidation is SO₂, which is observed in the flue gas. The NO_x removal of the spent photocatalyst was inhibited by 40%. The NO₂ selectivity increased by two-and-a-half times compared to the fresh photocatalyst. This research group investigated photo-SCR on real flue gas after the combustion of natural gas. The flue gas was taken from a 60 kW_th combustion chamber. In the first short-term test, when a fresh catalyst was used (see Figure 4), the deNO_x efficiency was ~80%. Nevertheless, the deNO_x efficiency varied in the range of 10–36% when the spent catalyst was applied and propane-butane was applied as fuel in the combustion chamber.

2.2. The Progress of Photo-Thermo-Catalytic Reactors for the NO_x Removal Process

The application of photo-thermo-catalytic reactors in the NO_x removal processes has not often been reported in the scientific literature. The main focus has been on the application of PTC reactors in the process of fuel generation (H₂, CH₄, and others), the degradation of organic pollutants (VOC and dyes), and organic synthesis. Early lab-scale designs of PTC reactors for NO_x removal were presented in 2009/2010 by Lasek and Wu. They applied the ideas of Poulston and colleagues [17]. The photoreactor was made of 6061 aluminum alloy and had a quartz window at the top. The shape of the reactor was a double isosceles triangle with a cut apex, and the size included a length of 35 mm, width of 12 mm, and depth of 1 mm. The light exposure area and internal volume of the reactor were 273 mm² and 0.546 mL, respectively. The temperature of the reactor was controlled by a hot plate equipped with a temperature controller. The temperature measurement inside the reactor was realized using a type K thermocouple that was placed in a drilled hole in the wall of the reactor (2 mm in diameter and 26 mm in length). The scheme of the lab-scale photo-thermo-rector and a photograph of this reactor are presented in Figure 5 and Figure 6. This reactor was successfully applied in the research of photo-thermo-SCR [3,19,20], photo-transformation of NO₂ into NO in the presence of N₂ [33], as well as the photo-epoxidation of propylene [34,35,36].

The next step in the development of PTC reactors in terms of NO_x removal was the scaling-up of the reactors. The research group was supervised by J. Lasek (Polish side) and J.C.S. Wu (Taiwanese side). They presented a new type of pilot-scale photo-reactor. The scheme and a photograph of this reactor are presented in Figure 7 and Figure 8. The pilot-scale photo-reactor consists of double quartz pipes (annular type, diameter of 100 mm and 80 mm, length of 1120 mm). The other parameters are as follows: rate of 50–300 L/h of air, 6 UVA lamps (PHILIPS TL 60 W/10R), electric power demand of 60 W, and UVA power of 15.8 W for each lamp. In this reactor, the photocatalysts on ball-shaped supports (diameter of ~3 mm) were mixed with cylinder-shaped silica glass for efficient light transportation into the photocatalyst surface. Moreover, thermal-assisted photocatalytic reactions can be carried out using this reactor, because reacting gases can be heated up to 150 °C. Currently, this reactor has been reconstructed. A new diode light source was applied, and the gas flow rate increased up to 5000 L/h.

2.3. The Importance of Light Transfer in Pilot-Scale Reactors

A very important issue in pilot- and real-scale photoreactors is light transfer from the light source to the photocatalyst surface. It is known that only the irradiated surface of the photocatalyst is available for an efficient photocatalytic reaction. In other words, the catalyst surface “must be visible” for the light source. Covered or shadowed surfaces are not efficient for photocatalytic reactions. Thus, the discovery of an efficient photocatalyst after lab-scale investigations is only “the halfway path” to a successful application. The second important issue is the efficient light transfer from the light source to the photocatalyst surface. In PTC, this task seems to be more difficult, because UV light sources are especially sensitive to temperature. It is known that some UV sources (e.g., diodes) exhibit sensitivity against temperature. Namely, the effect of temperature on UV LED degradation is described in the literature. For example, thermal exposure of GaN-based LED resulted in the degradation of optical power by 47% at an input current of 20 mA, after thermal exposure (i.e., 250 °C) for 160 h [38]. When considering a candidate for UV-light transmission/dispersion, the most common suggestion is silica glass. This material can be used for mechanical mixing with photocatalysts, but covering some photo-active surfaces with films made of silica glass is hard due to the high-temperature processing of silica glass (i.e., more than 1500 °C) [39]. Nevertheless, new UV-transparent materials are being developed. For example, silicone [40,41], polymethylmethacrylate (PMMA), and polycarbonate (PC) plastics [42,43] exhibit some transparency (more than 80%) for UV-A irradiation. The greatest advantage of these materials is their significantly lower processing temperature. However, the application of these materials in temperature-assisted photo-catalysis should be carefully considered due to the lower temperature of molding and service. Another candidate for efficient UV-A is low soda glass (such as Pyrex). This is an economy material that transmits UV-A (λ = 365 nm). Lukes et al. [44] as well as Shih et al. [45] investigated the optical transmittance of Pyrex glass. They observed that the transmittance of Pyrex glass (thickness of 2 mm) was higher than 85% in the wavelength range of 365–600 nm. Two important factors influence efficient light distribution in photo-thermo reactors, namely, UV-light loss affected by the glass (or another transparent material) barrier and the distance between the light source and the catalyst surface.

Thus, considering the above analysis, there are important issues to estimate the impact of distance between the light source and catalyst as well as the impact of barriers (i.e., transparent materials such as glass) on the UV light distribution. The impact of distance between the light source and catalyst surface is caused by the fact that the light is transmitted from the light source in a specific beam angle. For the tested diode in our reactor system, the specific beam angle was 160°. The light distribution curve of the UV-A (365 nm) 10 W diode is presented in Figure 9. The impact of distance on the UV light irradiation is presented in Figure 10. It is clear that due to the specific beam angle, the loss of light intensity is not linear. Moreover, it is important to estimate the loss in UV-light irradiation due to barriers. A comparison of this parameter for the selected materials and their thicknesses is presented in Figure 11.

Figure 12 shows an idea of the light transfer in a pilot-scale photo-reactor. The catalyst balls are mixed with quartz pipes to enhance light transfer into the catalyst surface. This idea was realized in the pilot-scale photo-reactor (see Figure 7 and Figure 8).

2.4. Beneficial Effects of Temperature in the Process of Photocatalytic NO_x Removal: Auxiliary Processes

Besides the main advantage of a temperature increase on photo-SCR (i.e., solving the “water problem” as, e.g., inhibited transformation of NO into NO₂), there are additional benefits due to such a combination of photo and thermal processes. The increase in temperature positively influences auxiliary processes that can be beneficial for photo-SCR.

Yamamoto and colleagues [46] reported a favorable effect of temperature in the photo-assisted selective catalytic reduction of NO_x (photo-SCR) in the presence of NH₃ as a reducing agent, over a TiO₂ photocatalyst and under UV light irradiation. It is known that the introduction of SO₂ (300 ppm) decreases the activity of the photo-SCR at 373 K. This undesired effect was inhibited by increasing the reaction temperature. In fact, increasing the reaction temperature increased the resistance to SO₂ gas, and at 553 K, NO conversion was stable for at least 300 min of reaction. Photo-SCR was carried out in a conventional fixed-bed flow reactor at atmospheric pressure. The light source was a 200 W Hg-Xe lamp equipped with fiber optics, a collective lens, and a mirror. The measured light irradiance was 360 mWcm⁻². The authors further explained that during the reaction in the presence of SO₂, an amorphous (NH₄)₂SO₄ species was formed between the TiO₂ particles, which inhibited the activity of the photo-SCR. The formed amorphous (NH₄)₂SO₄ species blocked the surface of the TiO₂, inhibiting photo-SCR. The deactivation is due to pore plugging caused by the deposition of the (NH₄)₂SO₄ species on the TiO₂ surface (see Figure 13). The amount of the (NH₄)₂SO₄ species decreased with an increased reaction temperature. This is an explanation for the positive effect of temperature on photo-SCR using NH₃ as a reducing agent and the presence of SO₂.

Another beneficial effect is the possibility of catalyst surface renovation during photo-thermal processes. Lasek and colleagues [19] revealed the potential self-renovation of a Pd/TiO₂ photocatalyst under higher temperatures. Indeed, during the photocatalytic reaction at room temperature and under oxidative conditions, palladium metal (as Pd⁰ state) is oxidized to PdO (palladium as Pd²⁺ state). This can be observed on a macro scale as a change in the catalyst color from dark to orange/beige. It is known that Pd-TiO₂ is more efficient than oxidized PdO/TiO₂. Therefore, the conversion of PdO to Pd during the photocatalytic process is beneficial in terms of NO_x removal. The authors observed that palladium oxide can be converted to palladium metal (as Pd⁰ state) via palladium oxide hydrate, palladium nitrate, and palladium acetate (as intermediates) under conditions of humidity, excess propane, and high temperature.

3. Further Development of Photo-Thermo-Catalytic Reactors

The application of a pilot-scale photo-reactor to the process of air purification from VOCs and NO_x is proposed as a development step in the PTC removal of gaseous pollutants. It is known that in some cases, VOC and NO_x exist together in places such as garages, tunnels, painting rooms, manufacturing buildings, and petrol stations. Thus, the potential applications of this technology are dedicated to such specific destinations. This topic seems to be much more challenging (compared to the application using flue gas) due to a much higher concentration of oxygen (i.e., ~21 v.% in the air). Our ongoing project under the Polish-Taiwanese/Taiwanese-Polish Joint Research Project focuses on such issues. Moreover, a new concept of a photo-catalytic reactor for honeycomb-shaped catalysts is suggested. The scheme of the reactor is presented below (see Figure 14). This reactor was designed by our group. It consists of a few chambers in which honeycomb catalysts can be placed. The challenge in such a type of reactor is in sufficient light being transported from the light source to the catalyst surface. UV diodes are applied as a light source, and the irradiation of the catalyst can be realized using quartz windows. The light sources are placed outside the reactor, thus allowing external heat into the reactor and cooling down the diodes that are necessary for such a technology of light.

The enhancement of photo-catalytic processes can be realized by the raising of temperature and pressure. Durga Devi and colleagues [47] noticed that increasing the temperature enhances the reaction kinetics, whereas an increase in pressure enhances the surface adsorption of gaseous substrates [47]. Higher pressure can bring beneficial effects due to the influence on the solubility of gas in the liquid phase [48,49] or the effect of pressure on the gas adsorption on solid states. For example, an increase in pressure enhances the surface adsorption of CO₂ [47].

Rehm observed an enhancement of chemical conversion by nearly 100% when the system pressure was increased to 20 bars [50]. Pressurized photo-reactors are mentioned in the literature [6,49,50,51,52,53,54]. An overview of the selected pressurized photo-reactors is given in

Table 3. Examples of pressurized photo-thermo-catalytic reactors.

Reactor	Process	Temperature	Pressure	Ref.
1.7 L, AISI 316 stainless	Photocatalytic reduction of CO2 with H2O	80 °C	7 bar	[57]
8 cm inner diameter and 1 cm depth	Photocatalytic reduction of CO2	75–95 °C	1.8 atm	[47]
Sapphire glass tube (o.d. 10 mm, length: 120 mm, wall thickness: 1 mm), filled with glass beads with a diameter of 6 mm	Dye-sensitized hydroxylation of arylboronic acid in an aqueous ethanol solution with high excess air compared to the substrate solution	n/a	1–20 bar	[50,51]
	CO2 photoreduction in the liquid phase	n/a	7 bar	[53]
Lab-scale photo-reactor	CO2 photoreduction in the liquid phase	90 °C	20 bar	[48,49]
Lab-scale (observation window ~1 cm2) pressurized photo-thermal reactor	DRIFT reactor for investigation of different gas–solid processes	Up to 450 °C	34.4 bar	[6]
Innovative pressurized photoreactor, 1.3 L	Photoreduction of CO2	90 °C	20 bar	[54]
Vortex reactor	Photo-oxidation of α-terpinene and furfuryl alcohol and photodeborylation of phenylboronic acid	80 °C	20 bar	[52]

. An example is the pressurized reactor reported by Rehm [50] (e.g., 1–20-bar pressurized upright-standing high-pressure-resistant sapphire glass tube reactor (o.d. 10 mm, length: 120 mm, wall thickness: 1 mm), which was filled with glass beads with a diameter of 6 mm), or the high-pressure (20 bars) photo-thermal (up to 90 °C) reactor for CO₂ photo-conversion to fuels reported by Rossetti and colleagues [49]. Keller and colleagues [6] presented a lab-scale (observation window ~1 cm²) pressurized photo-thermal reactor allowing for operation in up to 34.4 bars and 450 °C. Durga Devi and colleagues [47] presented a pressurized photo-reactor of 8 cm inner diameter and 1 cm depth. The reaction temperature and pressure were 75–95 °C and 1.8 atm, respectively. The reactor was applied in the process of the photocatalytic reduction of CO₂ into CH₄ over Pt-coated, graphene-oxide-wrapped TiO₂ nanotubes. Considering the impact of pressure on NO [55,56] and H₂O adsorption, it is proposed that the increase in pressure in the photo-thermo-catalytic process can bring a beneficial effect.

The Institute of Energy and Fuel Processing Technology designed and purchased a pressurized photo-thermo-catalytic reactor (supplied by Koncept-Tech, Gliwice, Poland) of 0.7 L volume and a flow rate in the range of 0.06–1.2 m_n³/h. The maximal temperature and pressure are 150 °C and 50 bars, respectively. This reactor is ready to be used in investigations of pressurized photo-thermo-catalytic NO_x conversion (oxidation or/and reduction). A photograph of the pressurized photo-thermal reactor is presented in Figure 15.

4. Conclusions

The development of photo-thermo-catalytic processes in terms of NO_x removal is presented in this study. An increase in temperature in photocatalytic systems to the moderate level of 100 °C provides beneficial effects for the photo-selective catalytic reduction of NO_x (photo-SCR). Namely, when a reducing agent occurs in the reaction system (e.g., C₃H₈), the increase in temperature inhibits the transformation of NO into NO₂ in the presence of oxidative compounds (such as O₂ and H₂O). Thus, the increase in temperature avoids such undesired effects for photo-SCR. This disclosure has helped to push photo-SCR from the lab scale to the pilot scale when the real condition of flue gas was applied. The presence of oxygen and water vapor in real-condition flue gas is obvious. Therefore, the presented progress in photo-SCR has increased the readiness level of the technology, and TRL using this method is now closer to commercial applications.